Ileal digestibility of amino acids in novel organic protein feedstuffs for pigs: Black soldier fly larvae meal (Hermetia illucens) Tiina Kortelainen Hilkka Siljander-Rasi Mikko Tuori Kirsi Partanen Ileal digestibility of amino acids in novel organic protein feedstuffs for pigs: Black soldier fly larvae meal (Hermetia illucens) Tiina Kortelainen Hilkka Siljander-Rasi Mikko Tuori Kirsi Partanen December 2014 3 Uusien luomuvalkuaisrehujen aminohappojen sulavuus sioilla: Mustasotilaskärpäsen toukkajauho (Hermetia illucens) Tiina Kortelainen 1) , Hilkka Siljander-Rasi 1) , Mikko Tuori 1) , Kirsi Partanen 2) 1) MTT Kotieläintuotannon tutkimus, Metla, PL 18, 01301 Vantaa, Finland, etunimi.sukunimi@mtt.fi 2) Snellmanin Lihanjalostus Oy, Kuusisaarentie 1, 68600 Pietarsaari, Finland Tiivistelmä Tutkimuksen tavoitteena oli määrittää uuden, luonnonmukaisesti tuotetun eläinperäisen rehuai- neen, mustasotilaskärpäsen toukista valmistetun jauhon (Hermetia illucens) aminohappojen standardoitu ohutsuolisulavuus kasvavilla porsailla. Toukkajauhon käyttö sikojen ruokinnassa ei toistaiseksi ole sallittua, mutta käytön mahdollisuutta tullaan käsittelemään EU-lainsäädännössä. Sveitsistä (FiBL Research Institute of Organic Agriculture) tuli kaksi toukkaerää, joista ensim- mäisessä oli erotettu rasvaa mekaanisesti, toisessa heksaaniuutolla. Kokeessa oli 40 porsasta, 17 imisää ja 23 leikkoa, jotka painoivat kokeen alussa keskimäärin 17 kg. Porsaat olivat duroc-, maatiais- ja yorkshirerotujen risteytyksiä. Totutusjaksoilla siat saivat luomuporsasrehua vapaasti porsimiskarsinassa (koejakso 0) ja samaa luomurehua siirryttyään lihasikalaan kahden porsaan karsinoihin (koejakso 1). Aminohappojen ohutsuolisulavuuden määritystä varten (koejakso 2) siat siirrettiin tärkkelyspohjaiselle rehulle. Koeruokintoja oli viisi: 1) vähäproteiininen rehu ami- nohappojen endogeenisen perustason erityksen määritystä varten, 2) rehu, jossa oli erän 1toukkajauhoa 10,2 %, 3) rehu, jossa oli erän 1 toukkajauhoa 20,4 %, 4) rehu, jossa oli erän 2 toukkajauhoa 9,3 % ja 5) rehu, jossa oli erän 2 toukkajauhoa 18,6 % rehun kuiva-aineessa (KA). Ryhmien 2‒5 rehuissa oli valkuaisen lähteenä myös heravalkuaisjauhetta (WPC) 22,85 % rehun kuiva-aineessa. Jokaisessa ryhmässä oli 8 porsasta pariruokinnalla. Teurastus tapahtui porraste- tusti ja ohutsuolen sisältöä kerättiin 50 ‒ 60 cm:n matkalta aminohappojen ohutsuolisulavuuden määrittämistä varten. Maksa, munuaiset ja tyhjä mahalaukku punnittiin ja mahahaavojen esiin- tyminen arvioitiin. Toukkajauho, josta rasvaa oli poistettu mekaanisesti (erä 1), sisälsi 629 g raakavalkuaista, 185 g raakarasvaa ja 51 g tuhkaa/kg KA. Vastaavat arvot toukkajauholle, josta rasvaa oli poistettu hek- saaniuutolla (erä 2), olivat 705 g, 90 g ja 53 g/kg KA. Ensimmäisen erän toukkajauhossa oli 31,7 g lysiiniä, 12,0 g metioniinia, 3,5 g kystiiniä ja 39,6 g/kg KA valiinia. Vastaavat arvot toisen erän toukkajauholle olivat 37,8 g, 14,1 g, 3,7 g ja 44,2 g/kg KA. Toukkaerä, toukkajauhon lisäystaso tai porsaiden sukupuoli eivät vaikuttaneet dieettien amino- happojen näennäiseen ohutsuolisulavuuteen. 4 Mekaanisesti erotetun (erä1) toukkajauhon näennäiset (AID) ja standardoidut (SID) aminohap- pojen ohutsuolisulavuudet olivat suurempia verrattuna heksaaniuutettuun (erä 2) toukkajauhoon. Välttämättömien aminohappojen AID oli ensimmäisessä toukkajauhoerässä 79,3 ‒ 93,2 % ja toi- sessa toukkajauhoerässä 61,2 ‒ 79,9 %. Välttämättömien aminohappojen SID oli ensimmäisessä toukkajauhoerässä 81,3 ‒ 94,8 % ja toisessa toukkajauhoerässä 64,0 – 81,8 %. Lysiinin standar- doitu ohutsuolisulavuus oli ensimmäisessä toukkajauhoerässä 81,3 %, metioniinin 90,7 %, kys- tiinin 49,8 %, treoniinin 82,5 % ja valiinin 92,9 %. Vastaavat sulavuudet toisessa toukkajau- hoerässä olivat 77,2 %, 81,8 %, -10,8 %, 64,0 % ja 73,6 %. Lähes kaikilla (87,5 %:lla) vähäproteiinista rehua saaneilla porsailla oli vakava mahahaava, joka todennäköisesti aiheuttaa kipua porsaalle. Toukkajauhoa saaneista porsaista 75 ‒ 100 %:lla ei ol- lut mahahaavaa, tai mahan limakalvomuutokset olivat vain vähäisiä. Vain alle kolmasosalla toukkaryhmien porsaista oli vakavia mahahaavoja, eikä yhtään pahimmanlaatuista mahahaava havaittu. Porsaiden maksan ja munuaisten paino ja munuaisten paino suhteessa elopainoon kas- voivat, kun toukkajauhoa lisättiin rehuun. Tulokset osoittavat, että rasvan erotusmenetelmä vaikuttaa mustasotilaskärpäsen toukista tehdyn jauhon aminohappojen sulavuuteen, sillä heksaaniuutetun toukkajauhon sulavuusarvot olivat huonommat kuin mekaanisesti erotetun toukkajauhon. Mustasotilaskärpäsen toukkajauho sisältää runsaasti hyvin sulavia aminohappoja, joten sen avulla voisi parantaa luomuporsasrehujen ami- nohappotasapainoa. Toukkajauho voisi monipuolistaa valkuaisen lähteitä sikojen luomuruokin- nassa, mutta sen tuotannon taloudellisia näkökulmia tulisi selvittää. Ohutsuolisulavuuden määri- tyksessä käytetyt hienojakoiset, tärkkelyspohjaiset rehut, erityisesti aminohappojen endogeenisen erityksen määrityksessä käytetty vähäproteiininen rehu, aiheuttivat porsaille mahahaavoja. Tut- kimusmenetelmiä tulisi kehittää niin, että niistä aiheutuisi mahdollisimman vähän haittoja eläin- ten terveydelle. Avainsanat: Sika Porsas Sulavuus Ohutsuolisulavuus Aminohapot Mustasotilaskärpänen Hermetia illucens Toukka 5 Ileal digestibility of amino acids in novel or- ganic protein feedstuffs: Black soldier fly larvae meal (Hermetia illucens) Tiina Kortelainen 1) , Hilkka Siljander-Rasi 1) , Mikko Tuori 1) , Kirsi Partanen 2) , 1) MTT Agrifood Research Finland, Animal Production Research, Metla, PO Box 18, 01301 Vantaa, Finland, firstname.lastname@mtt.fi 2) Snellmans Köttförädling Ab, Kuusisaarentie 1, 68600 Pietarsaari, Finland Abstract The objective of this study was to determine the standardised ileal digestibility (SID) of amino acids in organically produced black soldier fly larvae (Hermeti illucens) meal in growing piglets. The use of Hermetia meal in pig feeding is not allowed for the time being, but feed legislation in the EU concerning the use of Hermetia meal for pigs is in progress. Two batches of Hermetia meal arrived from Switzerland (FiBL Research Institute of Organic Agriculture). In batch 1, fat was extracted by mechanical extraction and in batch 2 hexane ex- traction was used.The experiment was carried out with a total of 40 growing piglets, 17 gilts and 23 barrows, with the initial body weight of ca. 17 kg. The piglets were distributed in experi- mental groups from litters of Finnish Landrace or Finnish Yorkshire x Finnish Landrace sows in- seminated with mixed semen from Duroc and Norwegian Landrace crossbred boars. Piglets were first fed in the farrowing pen with organic feed for piglets (period 0). The piglets received the same diet when they were moved to the fattening unit (2piglets/pen) (period 1). Diets were switched to starch based diets for the determination of the SID of amino acids. There were five dietary treatments: 1) low-protein diet to determine basal endogenous losses of amino acids, 2) diet with batch 1 Hermetia meal 10.2%, 3) diet with batch 1 Hermetia meal 20.4%, 4) diet with batch 2 Hermetia meal 9.3%, and 5) diet with batch 2 Hermetia meal 18.6% (of diet DM). Diets in groups 2‒5 contained also 22.85% (of diet DM) whey protein concentrate (WPC) as a protein source. There were 8 pigs per treatment in pair feeding. At the end of the trial, 3.5 h after the morning feeding, the piglets were stunned by bolt pistol, bled and ileal digesta was collected for digestibility determination. Liver, kidneys and empty stomach was weighed and stomach was visually estimated for gastric ulcers. Hermetia meal batch 1(mechanical fat extraction) contained 629 g crude protein, 185 g crude fat and 51 g ash per kg DM. Corresponding values for Hermetia meal in batch 2 (hexane fat extrac- tion) were 705 g, 90 g and 53 g/kg DM, respectively. There was 31.7 g of lysine, 12.0 g of me- thionine, 3.5 g of cystine and 39.6 g of valine per kg diet DM in Hermetia meal batch 1. Corre- sponding values for Hermetia meal batch 2 were 37.8 g, 14.1 g, 3.7 g and 44.2 g/kg DM, respec- tively. 6 There were no differences in the apparent ileal digestibility (AID) of amino acids in the experi- mental diets between the Hermetia meal inclusion levels or batches. The sex of the piglets did not affect the AID of amino acids in the experimental diets. The AID and the SID of the amino acids was higher in Hermetia meal batch 1 compared to Hermetia meal batch 2. The AID of essential amino acids varied between 79.3‒93.2% in batch 1 and 61.2‒79.9% in batch 2. The SID of essential amino acids varied between 81.3‒94.8% in batch 1 and 64.0‒81.8% in batch 2. The SID of lysine in Hermetia meal batch 1 was 81.3%, me- thionine 90.7%, cystine 49.8%, threonine 82.5% and valine 92.9%. Corresponding values for Hermetia meal batch 2 were 77.2%, 81.8%, -10.8%, 64.0% and 73.6%. Most of the piglets (87.5%) fed low-protein diet had severe gastric lesions in the oesophageal ar- ea (grades 2 and 3) which are expected to cause pain and reduce the welfare of the piglets. In di- ets with Hermetia meal 75‒100% of the piglets had no gastric lesions or the lesions were only minor. Severe gastric lesions were found in less than one third of the piglets fed with Hermetia meal and no grade 3 lesions were found. Feeding Hermetia meal to piglets increased the size of liver and kidneys and the proportion of kidneys in relation to live weight. Results indicate that the fat extraction method in Hermetia meal affects the AID and the SID of amino acids, as the digestibility values were lower in hexane extracted Hermetia meal compared to mechanically extracted Hermetia meal. Hermetia meal provides highly digestible amino acids, which can improve the amino acid balance in organic feeds for piglets. Hermetia meal could di- versify the protein supply for organic pig production, but the economic aspects of the production of Hermetia meal for pig feeding need to be explored. The fine-grained starch based feeds, espe- cially the low-protein feed used for the determination of the basal endogenous losses of amino acids, caused gastric ulcers for the piglets. The research methods should be developed to mini- mize the disadvantages to animal welfare. Keywords: Pig Piglets Digestibility Apparent ileal digestibility Standardised ileal digestibility Amino acids Black soldier fly Hermetia illucens Larvae 7 Table of Contents 1 Introduction ............................................................................................................................................... 8 1.1 Background ..........................................................................................................................................8 1.2 Objectives.............................................................................................................................................8 2 Materials and methods............................................................................................................................... 9 2.1 Test feedstuffs ......................................................................................................................................9 2.2 Animals and housing ..........................................................................................................................11 2.3 Experimental treatments.....................................................................................................................11 2.4 Diets and feeding ...............................................................................................................................12 2.5 Slaughter of pigs and collection of ileal digesta ................................................................................14 2.6 Calculation of digestibility and statistical analyses ............................................................................15 3 Results ..................................................................................................................................................... 16 3.1 Chemical composition of the experimental diets and feedstuffs ........................................................16 3.2 Apparent and standardised ileal digestibility of amino acids .............................................................18 3.3 Gastric health and organ weight .........................................................................................................22 4 Discussion and conclusion ...................................................................................................................... 26 5 References ............................................................................................................................................... 28 6 Appendix ................................................................................................................................................. 29 I References for analytical methods used .................................................................................................29 II The scale for the evaluation of the gastric lesions in pigs ....................................................................31 III Photograph of dry Hermetia meal .......................................................................................................32 8 1 Introduction 1.1 Background Currently, there are many challenges in food production: growing world population, over con- suming, climate changes and running out of non-renewable natural resources. Meat consumption is increasing, which increases the demand for new protein sources in livestock feeding. Insects are already used in feeding pets like iguanas and birds and insects are used as human food in many countries. Legislation does not yet allow feeding livestock with insect based processed an- imal protein (PAP). Insect fat anyhow, is allowed to be fed to non-ruminants. There is no specif- ic section for “insect meal” in Catalogue of Feed Materials (EC 68/2013) although there is listing for “whole or parts of terrestrial invertebrates” suggesting the use of insect derived protein may be possible. There are still many regulations to be met if insects would be used as feed. These regulations concern for example maximum permitted levels of contaminants such as heavy met- als, and processing practises for turning insects to PAP. But due to BSE outbreak regulations (EC 999/2001) prohibited all PAP (hydrolysed protein is exception) from being used in animal feed. Exception was also made in 2013 for fishmeal (non-ruminant PAP use for farmed fish spe- cies intended for human consumption). This regulation concerns slaughterhouse processing and because insects do not have similar slaughtering processes, insect PAP is forbidden. Insects reared for the production of PAP would consider being “farmed animals” and with current legis- lation this causes problems concerning the material the insects are grown on. Waste products from bioethanol production (wheat protein and barley hulls) are found in the Catalogue of feed materials (EC 68/2013), and could therefore be used in insect feeding. Manure on the other hand is in the category of “not allowed to be used as feed for farm animals”. Housefly larvaes can re- duce manure substrate mass by 60% over a 10 day period, which would be a benefit for the envi- ronment as well. Vegetable and domestic waste and manure would be ecologically feasible waste to use in the feeding of insects (Koeleman 2014). Using insect PAP for livestock feeding is cur- rently under discussion in EU. Both house fly pupae and soldier fly larvae contain high quality protein which will support nor- mal growth of chicks. Nearly half of the larvae is crude protein and one third is fat. Some of the crude protein in larvae is in the form of chitin, which is unavailable for the monogastric animals (Partanen 2012). Soldier fly larvae contain only 70% as much protein as house fly pupae, but soldier fly larvae contain 265% as much fat as house fly pupae. This appears to be helpful in converting animal manure into quality feedstuff. Soldier fly is also easier to harvest than house fly or pupae due to its size (Newton et al. 1977). Newton et al. (1977) suggest that Hermetia meal is suitable ingredient in swine diets, especially because of its amino acid, ether extract and calcium content. They suggest that the inclusion level of 33% which they used might be too high, and that Hermetia meal could be better utilized when used at lower inclusion levels for pigs. 1.2 Objectives The objective of this study was to determine the standardised ileal digestibility (SID) of amino acids in black soldier fly (Hermetia illucens) meal in growing piglets. 9 2 Materials and methods 2.1 Test feedstuffs The SID of amino acids in new organic protein feedstuff of animal origin, black soldier fly (Hermetia illucens) meal, in growing piglets was studied. Two batches of Hermetia meal arrived from FiBL (Research Institute of Organic Agriculture, Switzerland) in April 2013. Salmonella test was done before the experiment and the result was negative. The photograph of dry Hermet- ia meal is presented in Appendix III. In batch 1, fat was extracted by mechanical extraction and in batch 2 hexane extraction was used. The second batch was not premilled for fat extraction, so it was ground at MTT’s laboratory through 2mm sieve. The chemical and mineral composition of Hermetia meal analysed for the ICOPP database of organic feedstuffs is presented in Tables 1‒3 (Kyntäjä et al. 2014). Table 1. Chemical composition of Hermetia pupae and meal. Hermetia pupae ´MBM´ 1 dried, full fat Hermetia meal ´MBM´ 1 defatted Hermetia pupae ´CHO´ 2 dried, full fat Hermetia meal ´CHO´ 2 defatted Hermetia pupae ´kitchen waste´ 3 full fat Hermetia meal ´kitchen waste´ 3 defatted Dry matter % 91.8 92.5 86.5 89.5 88.0 90.3 Ash g/kg DM 124 168 40 60 88 137 Crude protein g/kg DM 446 627 482 710 405 659 Crude fat g/kg DM 342 43 396 41 410 20 Crude fibre g/kg DM 102 130 112 183 112 172 Sugars g/kg DM 8 11 8 12 10 15 NDF g/kg DM 191 259 212 325 151 235 ADF g/kg DM 88 93 101 199 91 142 Lignin g/kg DM 18 0 50 29 13 23 ADF-N 4 g/kg DM 7 10 8 13 7 12 1 MBM = grown on meat and bone meal 2 CHO = grown on carbohydrate rich material 3 Kitchen waste = grown on kitchen waste 41ADF-N = The nitrogen in this form may be unavailable to the animal. There were no large differences in crude protein (CP) content between the Hermetia grown on different materials. When full fat samples were compared, the Hermetia grown on carbohydrate rich material had the highest crude protein content, and the Hermetia grown on kitchen waste had the lowest content of crude protein. There was more crude protein in defatted samples and the defatted Hermetia grown on carbohydrate rich material had the highest crude protein content, and the defatted Hermetia grown on meat and bone meal had the lowest content of crude protein. Analyses of crude fibre, NDF, ADF, ADF-N and lignin (Table 1) do not necessarily describe correctly the cell wall content in insects, because instead of cellulose, the main cell wall material in insects is chitin. However, cellulose and chitin have very similar molecular structures, except that cellulose contains hydroxyl groups and chitin contains acetamides at the C2 position of the monomers. They have different physicochemical properties, such as different solubility in di- verse solvents, but similar functions. Main function is to support cell and body surfaces. Chitin strengthens fungal cell walls and exoskeletons of arthropods, whereas cellulose strengthens the 10 cell wall of plant cells (Merzendorfer 2006). The ADF-N analyse was done in order to find out how much of the total nitrogen is protein and how much of the nitrogen comes from chitin. Ac- cording to Finke (2007), ADF fraction in insects contains significant amount of amino acids (9.3‒32.7%) of the ADF (by weight). Thus using ADF as a measure of chitin in insects results in an overestimation of chitin content of insects. Hermetia does not contain much of the sulphur containing amino acids (Table 2). Table 2. Chemical composition of Hermetia pupae and meal. Hermetia pupae ´MBM´ 1 dried, full fat Hermetia meal ´MBM´ 1 defatted Hermetia pupae ´CHO´ 2 dried, full fat Hermetia meal ´CHO´ 2 defatted Hermetia pupae ´kitchen waste´ 3 full fat Hermetia meal ´kitchen waste´ 3 defatted Amino acids Essential Arginine g/16 g N 4.8 4.9 4.5 4.7 5.2 4.9 Histidine g/16 g N 2.8 2.9 2.7 2.9 3.0 3.0 Isoleucine g/16 g N 4.3 4.4 4.2 4.4 4.5 4.3 Leucine g/16 g N 6.9 7.0 6.8 7.1 7.3 6.9 Lysine g/16 g N 5.1 5.3 4.9 5.1 5.7 5.4 Methionine g/16 g N 2.0 1.9 1.8 1.8 2.0 1.8 Phenylalanine g/16 g N 4.1 4.2 3.8 4.0 4.1 4.1 Threonine g/16 g N 3.9 3.9 3.7 3.9 4.0 3.9 Valine g/16 g N 6.0 6.2 6.0 6.4 6.3 6.2 Non-essential Alanine g/16 g N 5.9 6.0 5.9 6.2 6.3 6.2 Aspartic acid g/16 g N 9.4 9.8 8.7 9.2 9.9 9.5 Cystine g/16 g N 0.5 0.5 0.5 0.5 0.5 0.5 Glutamic acid g/16 g N 9.7 10.1 9.1 9.8 10.7 10.5 Glycine g/16 g N 5.9 6.1 6.0 6.4 6.5 6.3 Proline g/16 g N 5.3 5.4 5.4 5.7 5.8 5.7 Serine g/16 g N 3.9 4.1 3.9 4.2 4.3 4.3 Tyrosine g/16 g N 5.9 6.3 5.9 6.3 6.9 6.8 In vitro ileal digestibility DM % 85.0 78.7 82.9 71.9 87.2 77.8 N % 78.4 78.4 77.0 74.9 81.7 80.5 In vitro total tract digestibility OM % 84.8 77.4 82.8 73.8 86.8 77.4 1MBM = grown on meat and bone meal 2CHO = grown on carbohydrate rich material 3Kitchen waste = grown on kitchen waste According to Finke (2007), house fly pupaes and black soldier fly larvaes contain significant amounts of calsium in their cuticle, compared to other insects. The material on which the Her- metia is grown can affect the mineral content of Hermetia. In full fat samples (Table 3) the calsium content is much lower in Hermetia grown on carbohydrate rich material compared to Hermetia grown on meat and bone meal and kitchen waste (6.3, 37.1 and 24.0 g/kg DM, respec- tively). Also the content of iron is the lowest in Hermetia grown on carbohydrate rich material (Table 3). 11 Table 3. Mineral composition of Hermetia pupae and meal. Hermetia pupae ´MBM´ 1 dried, full fat Hermetia meal ´MBM´ 1 defatted Hermetia pupae ´CHO´ 2 dried, full fat Hermetia meal ´CHO´ 2 defatted Hermetia pupae ´kitchen waste´ 3 full fat Hermetia meal ´kitchen waste´ 3 defatted Minerals Calcium g/kg DM 37.1 52.9 6.3 10.1 24.0 38.4 Magnesium g/kg DM 2.7 3.6 2.3 3.5 2.3 3.5 Phosphorus g/kg DM 6.7 8.9 5.6 8.2 4.8 6.9 Sulphur g/kg DM 2.9 4.0 3.0 4.6 2.7 4.3 Potassium g/kg DM 6.9 7.7 7.1 8.6 6.5 8.2 Sodium g/kg DM 0.8 0.9 1.3 1.5 1.0 1.3 Fe mg/kg DM 97.8 172.4 74.4 118.9 155.6 302.1 Cu mg/kg DM 9.3 13.7 9.5 14.7 8.4 13.3 Zinc mg/kg DM 67.2 95.7 78.4 125.9 69.5 111.3 Manganese mg/kg DM 128.0 171.7 162.6 242.0 195.4 319.7 1 MBM = grown on meat and bone meal 2 CHO = grown on carbohydrate rich material 3 Kitchen waste = grown on kitchen waste 2.2 Animals and housing The experiment was carried out with a total of 40 growing piglets, 17 gilts and 23 barrows, with the initial body weight of ca.17.2 kg (std. 1.96). The piglets were distributed in experimental groups from litters of Finnish Landrace or Finnish Yorkshire x Finnish Landrace sows insemi- nated with mixed semen from Duroc x Norwegian Landrace crossbred boars. These sows were in an organic feeding experiment and the piglets were weaned at the age of 40‒49 days. Sows were taken away and the piglets stayed in their farrowing pen for 4 days. After that the piglets were moved to the fattening unit. In the fattening unit the piglets were housed 2 piglets per pen in pens of 0.8 m x 2.49 m. The pens had 1.63 m² of concrete floor and 0.77 m² of slatted metal floor (dunging area). The pen walls were made of concrete and there were vertical metal bars to sepa- rate pens at the dunging area. Piglets had a nose-to nose-contact with each other at the dunging area. Feed was given twice daily and water was available ad libitum from drinking nipples. Wood shavings were used as bedding material. 2.3 Experimental treatments A total of 40 piglets were randomly distributed to pens (2 piglets / pen) and the pens were ran- domly allotted to five dietary treatments (Table 4). Piglets in the same pen were not from the same litter. The experiment was in three periods, and there where different diets and different ob- jectives in these periods, but the same animals continued from start to finish. The purpose of pe- riod 0 was adaptation after weaning. Period 1 was used for adaptation to the new environment and new pen mate. The feed was the same organic feed as offered before. The purpose of period 2 was to determine the SID of amino acids in Hermetia meal by regression method. It was taken into accoutn that part of the nitrogen in insects is adhered in chitin and may not be available to the animal. The ADF-N was subtracted from the total nitrogen of the Hermetia meal and the di- ets were planned to contain the same amount of protein that was not originated from the ADF-N. 12 Table 4. The experimental treatments for the determination of the SID of amino acids (period 2) in organically produced Hermetia illucens meal (feed amounts are given on DM basis). Group Period 0 4 d Period 1 4-5 d Period 2 7 d (+ 2 d transition period) Piglets 1 – Low-protein Organic diet Organic diet Starch based, low-protein diet 8 2 - Hermetia meal Batch 1, Level 1 Organic diet Organic diet 89.8% starch based diet with 22.85% WPC + 10.2% Hermetia meal batch 1 8 3 - Hermetia meal Batch 1, Level 2 Organic diet Organic diet 79.6% starch based diet with 22.85%WPC + 20.4% Hermetia meal batch 1 8 4 - Hermetia meal Batch 2, Level 1 Organic diet Organic diet 90.7% starch based diet with 22.85% WPC + 9.3% Hermetia meal batch 2 8 5 - Hermetia meal Batch 2, Level 2 Organic diet Organic diet 81.4% starch based diet with 22.85% WPC + 18.6% Hermetia meal batch 2 8 WPC = whey protein concentrate Batch 1 = mechanical fat extraction Batch 2 = hexane fat extraction 2.4 Diets and feeding During the periods 0 and 1 piglets received organically produced feed for piglets. Dietary ingre- dients were oats (10%), wheat (32.3%), barley (10%), rapeseed expeller (11.3%), peas (24%), concentrate (12%) (RehuX-concentrate), and some minerals and salt. The starch based low- protein diet in group 1 in period 2 was similar to the low-protein diet that was used in the digest- ibility trial of the ICOPP project to determine the SID of amino acids in organic grass silage for pigs. The starch based diet contained the same raw materials as the starch based low-protein diet but the targeted CP content was 160 g/kg DM. This was achieved by increasing the amount of whey protein concentrate (WPC) and decreasing the amount of barley starch and some other mi- nor ingredients (Table 5). A portion of the starch based diet was replaced by Hermetia meal from either batch 1 or batch 2 according to the experimental group (Table 3.) The CP intake was still 160 g/kg DM. Starch based diet and Hermetia meal where weighed separately and mixed before feeding. Titanium dioxide was used as an indigestible marker in both of the period 2 feeds (3 g/kg DM). 13 Table 5. Dietary ingredients of the experimental diets. Period 2 Dietary ingredients, g/kg: Starch based, low-protein diet Starch based diet 1 Barley starch 767.4 620.1 WPC75 2 50.0 228.5 Sugar 80 80 Cellulose 30 30 Rapeseed oil 35 15 Monocalciumfosfate 16.1 13.2 Limestone 14.5 6.2 Mineral-vitamin mixture 3 4 4 Titanium dioxide 3 3 1 Used with Hermetia meal according to experimental group, see Table 3: In group 2: 898 g/kg DM starch based diet + 102 g/kg DM Hermetia meal batch 1 In group 3: 796 g/kg DM starch based diet + 204 g/kg DM Hermetia meal batch 1 In group 4: 907g/kg DM starch based diet + 93 g/kg DM Hermetia meal batch 2 In group 5: 814 g/kg DM starch based diet + 186 g/kg DM Hermetia meal batch 2 2 WPC = Whey protein concentrate 3 The organic mineral-vitamin mixture Sika-Hiven supplied per kg of feed: 0.44 g of Ca, 0.26 g of P, 0.18 g of di- gestible P, 40 mg of Mg, 36 mg of Na, 5.6 mg of Fe, 0.4 mg of Cu, 4 mg of Mn, 8.8 mg of Zn, 48 µg of I, 40 µg of Se, 1200 IU of vitamin A, 200 IU of vitamin D3, 10 mg of vitamin E 3a700, 9.08 mg as α-tocopherol, 60 µg of vit- amin K, 0.2 mg of vitamin B1, 0.4 mg of vitamin B2, 0.3 mg of vitamin B6, 10 µg of vitamin B12, 0.18 mg of biotin, 2.4 mg of niacin, 0.12 mg of folic acid, and 1.6 mg of pantothenic acid. The piglets were given 100 g DM per kg metabolic body weight (W 0.75 ), based on the initial body weight at the beginning of period 1. The daily allowance was increased by 200 g DM at the beginning of period 2. The piglets were fed twice daily. Basal feed and Hermetia meal were hand-weighed on daily basis and the daily ration was given to piglets in the afternoon (15:00) and morning (07:00) feeding. Water was added on top of feed. The piglets were gradually switched to starch based diets of period 2 at the end of period 1. This transition lasted for 2 days. Period 2 lasted for 7 days. Feeding was graded at the last morning and the piglets were slaughtered at the end of trial, 3.5 h after the morning feeding for the collec- tion of ileal digesta. The feeding in periods 0, 1 and 2 and the experimental procedures are sum- marised in Table 6. All animals had exactly the same procedures, but for half of the animals the period 1 lasted for five days and for half of the animals it lasted for four days. Reason for this was that it was not possible to slaughter 30 piglets at the same day. Consequently, first 10 ani- mals started the trial one week earlier than the others and were slaughtered at 3 th of June (Table 6), 20 animals were slaughtered at 10 th of June and the rest of the animals were slaughtered one day after that. Sub-samples of feeds and experimental feedstuffs were collected during the last 3 days of period 2, pooled per treatment for the analyses of proximate composition, amino acids and markers. 14 Table 6. The experimental procedures of periods 0, 1 and 2. Day Period Diets and procedures Unit piglets/ pen Date 1 Fri Period 0, adaptation to dry feeding Organic diet, weighing of the piglets was done day before farrowing pen litter 17.5.2013 2 Sat Period 0, adaptation to dry feeding Organic diet farrowing pen litter 18.5.2013 3 Sun Period 0, adaptation to dry feeding Organic diet farrowing pen litter 19.5.2013 4 Mon Period 0, adaptation to dry feeding Organic diet, weighing of the piglets farrowing pen litter 20.5.2013 5 Tue Period 1, adaptation to fat- tening unit and pen mate Organic diet fattening unit 2piglets/ pen 21.5.2013 6 Wed Period 1, adaptation to fat- tening unit and pen mate Organic diet fattening unit 2piglets/ pen 22.5.2013 7 Thu Period 1, adaptation to fat- tening unit and pen mate Organic diet fattening unit 2piglets/ pen 23.5.2013 8 Fri Period 1, adaptation to fat- tening unit and pen mate Organic diet fattening unit 2piglets/ pen 24.5.2013 9 Sat Gradual change to period 2 feeding Starch based low-protein diet for the group 1 and fattening unit 2piglets/ pen 25.5.2013 10 Sun Gradual change to period 2 feeding starch based diet with WPC + Hermetia meal for groups 2-5 fattening unit 2piglets/ pen 26.5.2013 11 Mon Period 2, adaptation to starch based diets Weighing of the piglets and increasing of daily ratio fattening unit 2piglets/ pen 27.5.2013 12 Tue Period 2, adaptation to starch based diets fattening unit 2piglets/ pen 28.5.2013 13 Wed Period 2, adaptation to starch based diets fattening unit 2piglets/ pen 29.5.2013 14 Thu Period 2, adaptation to starch based diets fattening unit 2piglets/ pen 30.5.2013 15 Fri Period 2, adaptation to starch based diets Removal of bedding material fattening unit 2piglets/ pen 31.5.2013 16 Sat Period 2, adaptation to starch based diets fattening unit 2piglets/ pen 1.6.2013 17 Sun Period 2, adaptation to starch based diets Staggering of feedings according to the slaughter schedule fattening unit 2piglets/ pen 2.6.2013 18 Mon Period 2, adaptation to starch based diets Staggering of feedings according to the slaughter schedule, weighing of the piglets, slaughtering of the pig- lets 3.5h after morning feeding fattening unit 2piglets/ pen 3.6.2013 2.5 Slaughter of pigs and collection of ileal digesta At the end of the trial, the piglets were stunned by bolt gun, bled and ileal digesta was collected for digestibility determination. After slaughter, the abdominal cavity was opened, and a 0.5‒0.6m piece of ileum, backwards from ileo-caecal junction, was isolated. Digesta was collect- ed from the isolated intestine and if that part of ileum was empty additional 0.5‒0.6m piece was isolated and digesta was collected. The digesta was frozen immediately. Digesta samples were freeze-dried and analysed for DM, ash, amino acids and markers. All the analytical methods used are presented in Appendix I. 15 Liver and kidneys of the piglets were removed, weighed and examined visually for abnormali- ties. The stomach of the piglets was opened, emptied and the weight of empty stomach was measured. The oesophageal area of the stomach was examined for gastric ulceration according to Hautala and Rautiainen (1991) (Appendix II). The consistency of digesta in stomach was evalu- ated as follows: Grade 1 = liquid, Grade 2 = liquid with visible particles, Grade 3 = mushy. 2.6 Calculation of digestibility and statistical analyses AID of amino acids were calculated by marker method as follows: AID, % = [1 - (AAdigesta/AAdiet) x (Mdiet/Mdigesta)] AA = amino acid, and M = marker concentration, g/kg DM Basal ileal endogenous losses of amino acids (IAAend), g/ kg DM intake, were calculated from ileal samples of low-protein diet according to Stein et al. (2007) as follows: Basal IAAend = AAdigesta × (Mdiet/Mdigesta) The standardised ileal digestibility of amino acids was calculated according to Stein et al. (2007) as follows: SID, % = AID + [(basal IAAend / AAdiet) × 100]. The AID and the SID of amino acids in the two different Hermetia meals and in the basal feed was calculated by regression method (Fan and Sauer 1995) using the GLM procedure of SAS® for Windows (Version 9.2).The AID of diets was analysed with SAS® for Windows (Version 9.2) using the MIXED procedure with fixed effects of Hermetia batch, inclusion level and sex of the piglets and interactions of Hermetia batch and level, Hermetia batch and sex, and inclu- sion level and sex. Organ weight data was analysed with SAS® for Windows (Version 9.2) with fixed effects of treatment, sex of the piglets and interaction of treatment and sex. 16 3 Results 3.1 Chemical composition of the experimental diets and feedstuffs The diet in periods 0 and 1 contained DM 877 g/kg, ash 58 g/kg DM, crude protein 177 g/kg DM, crude fibre 53 g/kg DM, lysine 8.5 g/kg DM, threonine 7.3 g/kg DM, and methionine + cys- tine 6.8 g/kg DM. The analysed chemical composition of the experimental diets in period 2 is presented in Table 7. Table 7. Analysed chemical composition of the experimental diets. Starch based low- protein diet Starch based diet 1 Analysed chemical composition, g/kg DM Dry matter g/kg 913 927 Ash 40 28 Crude protein 32 107 Amino acids, g/kg DM Essential Arginine 0.8 2.6 Histidine 0.9 2.4 Isoleucine 1.9 7.1 Leucine 3.4 12.4 Lysine 2.8 10.3 Methionine 1.3 3.2 Phenylalanine 1.2 3.9 Threonine 2.1 7.9 Valine 1.7 6.8 Non-essential Alanine 1.5 5.6 Aspartic acid 3.5 12.7 Cystine 0.6 2.1 Glutamic acid 5.4 20.0 Glycine 0.7 2.3 Proline 1.9 6.8 Serine 1.7 6.0 Tyrosine 0.9 3.2 1 Used with Hermetia meal according to experimental group, see Table 3. In group 2: 898 g/kg DM starch based diet + 102 g/kg DM Hermetia meal batch 1 In group 3: 796 g/kg DM starch based diet + 204 g/kg DM Hermetia meal batch 1 In group 4: 907g/kg DM starch based diet + 93 g/kg DM Hermetia meal batch 2 In group 5: 814 g/kg DM starch based diet + 186 g/kg DM Hermetia meal batch 2 17 The analysed chemical composition of the Hermetia meal is presented in Table 8. Mechanically extracted Hermetia meal in batch 1 contained more fat than hexane extracted Hermetia meal in batch 2. Mechanically extracted Hermetia meal contained a little less crude protein than hexane extracted Hermetia meal (629 and 705 g/kg DM, respectively) and therefore also less amino ac- ids. The content of glutamic acid in mechanically extracted Hermetia meal was much lower compared to hexane extracted Hermetia meal (64.7 and 73.0 g/kg DM, respectively). Mechanical extraction of fat leaves more fat to the Hermetia than hexane extraction (crude fat content 185 and 90 g/kg DM, respectively). Both batches contained 12 g ADF-N /kg DM. ADF-N was inter- preted to represent the part on nitrogen that is unavailable to the animal. Table 8. Analysed chemical composition of Hermetia meal. Hermetia meal mechanical fat extraction Hermetia meal hexane fat extraction Analysed chemical composition, g/kg DM Dry matter g/kg 925 906 Ash 51 53 Crude protein 629 705 Crude fat 185 90 NDF 318 287 ADF 136 143 ADF-N 1 12 12 Amino acids, g/kg DM Essential Arginine 30.5 32.0 Histidine 18.4 20.8 Isoleucine 26.9 30.8 Leucine 43.5 49.5 Lysine 31.7 37.8 Methionine 12.0 14.1 Phenylalanine 24.1 28.0 Threonine 24.3 27.4 Valine 39.6 44.2 Non-essential Alanine 39.6 45.6 Aspartic acid 57.4 63.5 Cystine 3.5 3.7 Glutamic acid 64.7 73.0 Glycine 34.3 40.4 Proline 36.3 41.2 Serine 27.1 30.1 Tyrosine 39.7 43.8 1 ADF-N = The nitrogen in this form may be unavailable to the animal. 18 3.2 Apparent and standardised ileal digestibility of amino acids The effect of Hermetia meal inclusion level and batch and sex of the piglets on the AID of amino acids in experimental diets is presented in Table 9.There were no differences in the AID of ami- no acids in the experimental diets between the inclusion levels or batches. The AID of cystine tended to be slightly higher at the lower inclusion level of Hermetia meal compared to higher in- clusion level (p=0.09). The sex of the piglets did not affect the AID of amino acids in the exper- imental diets. Table 9. The effect of Hermetia meal inclusion level and Hermetia meal batch and sex of the piglets on the AID of amino acids in experimental diets (%). SEM Hermetia batch SEM SEM Hermetia level 1 and 2 p level Mechanical fat extraction Hexane fat extraction p batch Sex Gilts Barrows p sex n 16 16 16 16 17 23 Amino Acid Essential Arginine 80.3 81.5 2.49 0.72 79.8 82.1 2.56 0.50 81.3 80.6 2.66 0.84 Histidine 80.7 78.7 2.61 0.58 77.9 81.5 2.69 0.33 79.9 79.5 2.80 0.91 Isoleucine 86.0 84.8 1.64 0.62 84.2 86.5 1.69 0.32 85.6 85.1 1.76 0.83 Leucine 86.6 85.5 1.73 0.65 84.8 87.4 1.78 0.29 86.5 85.7 1.85 0.74 Lysine 84.3 81.7 2.01 0.35 81.5 84.6 2.07 0.27 82.3 83.8 2.15 0.61 Methionine 87.2 85.8 1.74 0.56 85.2 87.9 1.79 0.28 86.3 86.7 1.86 0.89 Phenylalanine 83.5 83.8 2.14 0.92 82.1 85.3 2.20 0.29 84.7 82.6 2.28 0.48 Threonine 77.9 75.4 2.30 0.43 74.7 78.5 2.37 0.25 76.9 76.3 2.46 0.85 Valine 80.6 80.5 2.11 0.97 79.0 82.1 2.17 0.30 81.1 80.0 2.26 0.73 Non-essential Alanine 79.1 79.3 2.43 0.96 77.5 80.9 2.50 0.33 79.0 79.4 2.60 0.93 Aspartic acid 81.6 78.7 2.48 0.41 78.2 82.1 2.55 0.28 79.9 80.4 2.65 0.89 Cystine 73.0 62.1 4.60 0.09 63.4 71.7 4.74 0.21 68.0 67.1 4.92 0.89 Glutamic acid 81.3 76.7 3.14 0.29 76.8 81.3 3.23 0.31 78.0 80.1 3.35 0.63 Glycine 58.6 60.2 5.19 0.82 54.8 64.1 5.34 0.21 59.5 59.4 5.55 0.99 Proline 78.8 77.3 2.15 0.62 76.1 80.0 2.21 0.22 78.8 77.3 2.30 0.61 Serine 76.1 75.1 2.56 0.78 73.5 77.8 2.64 0.24 75.7 75.5 2.74 0.95 Tyrosine 77.8 79.4 3.31 0.72 76.0 81.1 3.41 0.28 81.7 75.5 3.54 0.20 n = number of animals SEM is the highest standard error 19 The interaction of the Hermetia meal inclusion level and Hermetia meal batch in the experi- mental diets is presented in Table 10. The AID of tyrosine was higher in diet with mechanically extracted Hermetia meal when the inclusion level was higher. In diet with hexane extracted Hermetia meal the AID of tyrosine was higher when the inclusion level was lower (p=0.01). Tendency to similar interaction was detected with arginine, isoleucine and valine. Table 10. The interaction of the two different Hermetia meal batches and inclusion levels and the effect on the apparent ileal digestibility of amino acids in the experimental diets (%). Hermetia, mechanical fat separation Hermetia, hexane fat extraction SEM p Level 1 Level 2 Level 1 Level 2 Batch x Level n 8 8 8 8 Essential Arginine 75.9 83.6 84.8 79.5 3.70 0.07 Histidine 76.3 79.5 85.1 77.9 3.89 0.17 Isoleucine 82.8 85.7 89.1 84.0 2.44 0.09 Leucine 83.2 86.3 90.0 84.7 2.58 0.10 Lysine 81.5 81.4 87.2 82.0 2.99 0.37 Methionine 84.5 85.9 90.0 85.8 2.58 0.26 Phenylalanine 79.4 84.7 87.6 82.9 3.17 0.11 Threonine 73.6 75.8 82.1 74.9 3.42 0.16 Valine 76.1 81.9 85.1 79.1 3.14 0.06 Non-essential Alanine 74.7 80.3 83.5 78.3 3.61 0.12 Aspartic acid 77.2 79.2 85.9 78.2 3.68 0.17 Cystine 64.5 62.3 81.5 61.9 6.84 0.19 Glutamic acid 77.0 76.6 85.7 76.9 4.66 0.34 Glycine 48.3 61.3 69.0 59.2 7.72 0.13 Proline 74.4 77.9 83.2 76.7 3.20 0.11 Serine 71.2 75.7 81.0 74.5 3.81 0.14 Tyrosine 68.9 83.1 86.6 75.7 4.93 0.01 n. = number of observations SEM is the highest standard error in Hermetia batch x Hermetia level The AID of amino acids in two Hermetia meal batches is presented in Table 11. The AID of amino acids in Hermetia meal and basal feed was calculated by regression method. The AID values were higher in mechanically extracted Hermetia meal compared to hexane extracted Hermetia meal. The AID of essential amino acids in mechanically extracted Hermetia meal was 79.3‒93.2% and 61.2‒79.9% in hexane extracted Hermetia meal. In mechanically extracted Hermetia meal the AID of cystine was very low (42.9%) and the results showed no reliable value for the AID of cystine in hexane extracted Hermetia meal as the value was negative. 20 Table 11. The apparent ileal digestibility of amino acids in two Hermetia meal batches. Hermetia meal mechanical fat extraction Hermetia meal hexane fat exraction Mean SEM Mean SEM n 16 16 Essential Arginine 92.9 10.71 77.2 7.96 Histidine 83.1 13.41 71.9 8.53 Isoleucine 91.5 11.26 75.2 8.42 Leucine 93.2 12.48 74.7 9.15 Lysine 79.4 16.81 75.6 10.22 Methionine 88.5 12.65 79.9 7.75 Phenylalanine 92.6 12.41 79.0 7.52 Threonine 79.3 19.38 61.2 11.73 Valine 91.5 11.73 72.4 8.88 Non-essential Alanine 88.8 12.55 74.7 8.65 Aspartic acid 81.9 16.50 67.9 10.28 Cystine 42.9 62.51 -17.3 28.44 Glutamic acid 73.1 25.43 63.0 15.00 Glycine 75.9 19.56 54.5 15.67 Proline 82.6 13.34 69.7 8.70 Serine 83.8 17.60 67.4 9.77 Tyrosine 99.7 13.51 64.2 11.4 n = number of animals SEM=standard error 21 The basal endogenous losses of amino acids (IAAend) of this experiment are presented in Table 12. For comparison, the mean values of basal ileal endogenous losses of amino acids from diets with casein/wheat gluten from the research of Jansman et al. (2002) are presented in Table 12. Table 12. Mean flow of basal ileal endogenous amino acids (g/kg DM intake). Hermetia meal experiment Jansman et al. (2002) 1 Mean Std. Mean n 8 Essential Arginine 0.46 0.106 0.36 Histidine 0.18 0.049 0.21 Isoleucine 0.38 0.107 0.51 Leucine 0.59 0.172 0.54 Lysine 0.51 0.165 0.44 Methionine 0.18 0.038 0.12 Phenylalanine 0.32 0.103 0.36 Threonine 0.65 0.152 0.72 Valine 0.50 0.133 0.74 Non-essential Alanine 0.55 0.127 0.56 Aspartic acid 0.90 0.236 0.95 Cystine 0.16 0.050 0.28 Glutamic acid 1.13 0.307 1.75 Glycine 1.10 0.468 0.70 Proline 2.34 1.506 0.76 Serine 0.61 0.127 0.91 Tyrosine 0.45 0.169 0.30 n = number of animals Std. = standard error 1 Basal ileal endogenous losses of amino acids from diets with casein/wheat gluten. The SID of amino acids is presented in Table 13. The SID of amino acids in Hermetia meal and basal feed was calculated by regression method. The SID of amino acids was higher in mechani- cally extracted Hermetia meal compared to hexane extracted Hermetia meal. The SID of essen- tial amino acids in mechanically extracted Hermetia meal was 81.3‒94.8% and 64.0‒81.8% in hexane extracted Hermetia meal. In mechanically extracted Hermetia meal the SID of cystine was very low (49.8%) and the results showed no reliable value for the SID of cystine in hexane extracted Hermetia meal as the value was negative. 22 Table 13. The standardised ileal digestibility of amino acids in two Hermetia meal batches. Hermetia meal mechanical fat extraction Hermetia meal hexane fat exraction Mean SEM Mean SEM n 16 16 Essential Arginine 94.3 10.71 78.6 7.96 Histidine 84.4 13.41 73.0 8.53 Isoleucine 93.2 11.26 76.7 8.42 Leucine 94.8 12.48 76.1 9.15 Lysine 81.3 16.81 77.2 10.22 Methionine 90.7 12.65 81.8 7.75 Phenylalanine 94.2 12.41 80.4 7.52 Threonine 82.5 19.38 64.0 11.73 Valine 92.9 11.73 73.6 8.88 Non-essential Alanine 90.4 12.55 76.1 8.65 Aspartic acid 83.7 16.50 59.6 10.28 Cystine 49.8 62.51 -10.8 28.44 Glutamic acid 75.3 25.43 64.9 15.00 Glycine 78.8 19.56 56.9 15.67 Proline 86.3 13.34 72.9 8.70 Serine 86.4 17.60 69.8 9.77 Tyrosine 100.7 13.51 65.1 11.40 n = number of animals SEM=standard error 3.3 Gastric health and organ weight Effect of feeding with starch based low-protein diet and four starch based diets with Hermetia meal on incidence and severity of gastric lesions and the results of visual evaluation of stomach contents are presented in Table 14. Most of the piglets (87.5%) fed low-protein diet had severe gastric lesions in the oesophageal area (grades 2 and 3) which are expected to cause pain and re- duce the welfare of the piglets (Hautala and Rautiainen 1991) (Appendix III). Their stomach contents were in liquid form with and without visible particles. In diets with Hermetia meal in which fat was extracted mechanically, 75‒87.5% of the piglets had no gastric lesions or the le- sions were so small that they should not affect the welfare of the piglets (grades 0 and 1). Corre- sponding values for Hermetia meal in which hexane fat extraction was used were 75‒100%. Se- vere gastric lesions were found in piglets fed with Hermetia meal but grade 3 lesions were not found. The stomach contents of piglets fed Hermetia meal were in liquid form but feed particles were also present. Two pigs fed diet with Hermetia meal, in which hexane fat extraction was used, had stomach contents in mushy form. 23 Table 14. The incidence and severity of gastric lesions and consistency of digesta in stomach of piglets fed starch based low-protein diet and diets containing Hermetia meal. Starch based diets Hermetia meal mechanical fat extraction Hermetia meal hexane fat extraction Low-protein Level 1 Level 2 Level 1 Level 2 n 1 % 2 n 1 % 2 n 1 % 2 n 1 % 2 n 1 % 2 Severity of gastric lesions Grade 0 0 1 12.50 0 1 12.50 1 12.50 Grade 1 1 12.50 6 75.00 6 75.00 7 87.50 5 62.50 Grade 2 3 37.50 1 12.50 2 25.00 0 2 25.00 Grade 3 4 50.00 0 0 0 0 Consistency of digesta in stomach 3 Grade 1 5 62.50 0 0 0 0 Grade 2 3 37.50 8 100.00 8 100.00 7 87.50 7 87.50 Grade 3 0 0 0 1 12.50 1 12.50 1 n=number of piglets within diet and grade. 2%=percentage distribution within diet and grade. 3 Consistency of digesta in stomach: Grade 1 = liquid, Grade 2 = liquid with visible particles, Grade 3 = mushy. The effect of diet on organ weights is shown in Table 15. The live weight was the lowest in the piglets fed low-protein diet before slaughter. Compared to piglets fed low-protein diet, the live weight was higher in piglets fed Hermetia meal but similar with each other. There were no dif- ferences in the weight of empty stomach between the experimental groups. The weight of empty stomach in relation to live weight was the lowest in piglets fed the diet with Hermetia meal in which fat was extracted mechanically, and it was significantly lower than that of piglets fed the low-protein diet. The weight of liver was the lowest in piglets fed low-protein diet. The weight of liver in relation to live weight did not differ between the experimental groups. The weight of kidneys and the weight of kidneys in relation to live weight increased when Hermetia meal was added to the diets. 24 Table 15. The effect of diet on the live weight and organ weight of piglets. Starch based diets Hermetia meal mechanical fat extraction Hermetia meal hexane fat extraction Low-protein Level 1 Level 2 Level 1 Level 2 SEM p diet n piglets 8 8 8 8 2 8 Live weight, kg 19.4 a 24.9 b 25.8 b 24.2 b 25.6 b 1.259 0.0015 Empty stomach, kg 1,2 0.147 0.157 0.168 0.173 0.172 0.01 0.227 Empty stomach,% of live weight 2 0.77 a 0.64 bc 0.65 bc 0.71 ac 0.67 ac 0.028 0.0085 Liver, kg 1 0.559 a 0.788 b 0.819 b 0.735 b 0.812 b 0.051 0.001 Liver, % of live weight 2.90 3.17 3.18 3.05 3.16 0.15 0.53 Kidneys, kg 1 0.078 a 0.142 b 0.150 b 0.135 b 0.146 b 0.009 <0.001 Kidneys, % of live weight 0.40 a 0.57 b 0.58 b 0.56 b 0.59 b 0.041 0.004 1 Stomach, liver and kidneys were weighed immediately after slaughter, 3.5 h after feeding. 2 One observation was rejected due to divergent value in empty stomach weight and empty stomach weight in relation to live weight in group 4 (Hermetia meal, hexane fat extraction, level 1). a,b,c Means with different superscript differ significantly (p<0.05). The effect of sex on the incidence of gastric lesions is shown in Table 16. Gilts had less severe gastric lesions than barrows (23.52% vs. 34.79%). The live weight, the weight of empty stom- ach, liver and kidneys and their proportion of live weight did not differ between gilts and bar- rows (Table 17). Table 16. The effect of sex on the incidence and severity of gastric lesions and consistency of digesta in stomach. Gilts Barrows n 1 % 2 n % Severity of gastric lesions Grade 0 2 11.76 1 4.35 Grade 1 11 64.71 14 60.87 Grade 2 2 11.76 6 26.09 Grade 3 2 11.76 2 8.70 Consistency of digesta in stomach 3 Grade 1 3 17.65 2 8.70 Grade 2 13 76.47 20 86.96 Grade 3 1 5.88 1 4.35 1 n=number of piglets within sex and grade. 2%=percentage distribution within sex and grade. 3 Consistency of digesta in stomach: Grade 1 = liquid, Grade 2 = liquid with visible particles, Grade 3 = mushy. 25 Table 17. The effect of sex on the live weight and organ weight of piglets. Gilts Barrows SEM p sex n piglets 17 23 1 Live weight, kg 24.4 23.5 0.78 0.39 Empty stomach, kg 1,2 0.165 0.162 0.006 0.73 Empty stomach, % of live weight 1 0.68 0.69 0.02 0.72 Liver, kg 2 0.753 0.732 0.031 0.62 Liver, % of live weight 3.10 3.08 0.10 0.92 Kidneys, kg 2 0.127 0.134 0.005 0.32 Kidneys, % of live weight 0.52 0.56 0.03 0.16 12 One observation was rejected due to divergent value in empty stomach weight and empty stomach weight in relation to live weight in group 4 (Hermetia meal, hexane fat extraction, level 1). 2 Stomach, liver and kidneys were weighed immediately after slaughter, 3.5 h after feeding. 26 4 Discussion and conclusion The content of crude protein in Hermetia meal in the present trial was comparable to feed table values for fish meal. The content of crude fat in hexane extracted Hermetia meal was compara- ble to feed table values for fish meal. In mechanically extracted Hermetia meal the content of crude fat was higher than in fish meal. The content of ash was lower in Hermetia meal compared to fish meal. The content of lysine and methionine in Hermetia meal was lower than the feed ta- ble values presented for fish meal. The content of threonine in Hermetia meal was close to that of fish meal. The content of cystine in Hermetia meal was clearly lower and the content of valine was clearly higher than in fishmeal (CVB 2011, EvaPig 2008). Estimated from the content of NDF, one third of the Hermetia meal is chitin. The AID and SID of amino acids were higher in mechanically extracted Hermetia meal (batch 1) compared to hexane extracted Hermetia meal (batch 2). The SID of essential amino acids in Hermetia meal in batch 1was 81.3‒94.8% and 64.0‒81.8% in batch 2. Hexane extraction may have affected the digestibility of amino acids in Hermetia meal. The SID of lysine and threonine in mechanically extracted Hermetia meal was somewhat smaller than presented in feed tables for fish meal, but the SID of arginine, methionine, isoleucine, va- line, leucine, phenylalanine and histidine was similar or even higher to that of fish meal (CVB 2011, EvaPig 2008). Compared to plant based protein sources, in mechanically extracted Her- metia meal the SID of nearly all essential amino acids, exept for threonine and histidine, was higher than in soybean meal. The SID of arginine was similar to soy bean meal. The SID of ami- no acids is generally higher in mechanically extracted Hermetia mea compared to peas (CVB 2011, EvaPig 2008). However, the SID of cystine is very low in Hermetia meal. Feeding starch based low-protein diet to determine basal endogenous losses of amino acids had very negative effect on gastric health of the piglets. After the seven day feeding period, seven of eight piglets fed low-protein diet had severe gastric ulcers which are expected to cause pain for the animals. Gastric ulcers in pigs can develop quickly, even in 12 hours, and healing can occur relatively quickly as well (Friendship 2004). Severe gastric ulcers can develop even in young pigs (Fossi et al. 2010). The piglets fed starch based diets with Hermetia meal had less gastric lesions and the most severe lesions were not found. The stomach contents of piglets fed low- protein diet were in liquid form with or without visible particles and the stomach contents of pig- lets fed Hermetia meal were mostly in liquid form with visible particles. Two piglets fed diet of mechanically extracted Hermetia meal had stomach contents in mushy form. According to Niel- sen and Ingvartsen (2000), gastric ulcers can be prevented by feeding factors which increase the firmness of stomach contents. The starch in the experimental diets was very fine-grained. The fi- ne feed structure and small particle size have been shown to increase the prevalence of gastric lesions in pigs (Mahan et al. 1966). In this experiment, feeding Hermetia meal to piglets increased the size of liver and kidneys and the proportion of kidneys in relation to live weight. It remains unclear whether there were harm- ful substances in Hermetia meal which could result to the increased size of kidneys and whether the rate of increase on the size of kidneys is detrimental for the animal. 27 In conclusion, in this experiment two fat extraction methods were compared and the extraction method affects the AID and the SID of amino acids, as the digestibility values were lower in hexane extracted Hermetia meal compared to mechanically extracted Hermetia meal. The results indicate that mechanically extracted Hermetia meal provides highly digestible amino acids, which can improve the amino acid balance in organic feeds for piglets. Hermetia meal could di- versify the protein supply for organic pig production, but the economic aspects of the production of Hermetia meal for pig feeding need to be explored. The fine-grained starch based feeds, espe- cially the low-protein feed used for the determination of the basal endogenous losses of amino acids, caused gastric ulcers for the piglets. The research methods should be developed to mini- mize the disadvantages to animal welfare. 28 5 References CVB 2011. Chemical compositions and nutritional values of feed materials. Centraal Veevoederbureau: Lelystad, The Netherlands. EC 68/2013 available at: http://eur- lex.europa.eu/LexUriServ/LexUriServ.do?uri=OJ:L:2013:029:0001:0064:EN:PDF pages 10, 17 and 48. EC 999/2001available at: http://ec.europa.eu/food/fs/afs/marktlab/marktlab14_en.pdf EvaPig 2008. EvaPig® Evaluation of Pig feeds. Energy, amino acids and phosphorus values for pigs. INRA, AFZ, Ajinomoto Eurolysine S.A.S. Cited 20.11.2014 at www.evapig.com Fan, M.Z., Sauer, W.C. 1995. Determination of Apparent ileal amino acid digestibility in peas for pigs with the direct, difference and regression method. Livestock Production Science 44:61−72. Finke, M.D. 2007. Estimate of Chitin in Raw Whole Insects. Zoo Biology 26: 105-115. Fossi, M., Karhapää, M., Partanen, K., Kortelainen T., Siljander-Rasi, H. 2010. Gastric ulcers in 7 8- week-old pigs, fed by various formulae of pelleted feed Proceedings the 2nd European Symposium on Porcine Health Management 27.-28.May 2010 : Hannover, Germany. Posters and abstracts. Edited by Elisabeth grosse Beilage, Thomas Blaha. Friendship, RM. 2004. Gastric ulceration in swine. Journal of swine health and production 12:34−35. Hautala, M., Rautiainen, E. 1991. A comparative study of feeder pig units using dry or liquid feeding in the western part of Finland. Part 2: Assessment and categorization of the gastric lesions in pigs. Suomen Eläinlääkärilehti 97:298−307. Jansman, AJM., Smink, W., van Leeuwen, P., Rademacher, M. 2002. Evaluation through literature data of the amount and amino acid composition of basal endogenous crude protein at the terminal ileum of pigs. Animal Feed Science and Technology 98:49–60. Koeleman, E. 2014. Insects crawling their way into feed regulation. All About Feed. 22 (6):18-21. Kyntäjä, S., Partanen, K., Siljander-Rasi, H., Jalava, T. 2014. Tables of composition and nutritional val- ues of organically produced feed materials for pigs and poultry. MTT Report 164 (2014). 37 p. ISBN 978-952-487-571-4. Available at: http://jukuri.mtt.fi/bitstream/handle/10024/484922/mttraportti164.pdf Mahan, DC., Pickett, RA., Perry, TW., Curtin, TM, Featherson, WR., Beeson, WM. 1966. Influence of various nutritional factors and physical form of feed on esophagogastric ulcers in swine. Journal of Ani- mal Science 25:1019−1023. Merzendorfer, H. 2006. Insect Chitin synthases: a review. Journal of Comparative Physiology B 176: 1– 15. Newton, G.L., Booram, C.V., Barker, R.W. and Hale, O.M. 1977. Dried Hermetia illucens Larvae Meal as a Supplement for Swine. Journal of Animal Science 44, 395-400. Nielsen, EK., Ingvartsen, KL. 2000. Effect of cereal type, disintegration method and pelleting on stomach content, weight and ulcers and performance in growing pigs. Livestock Production Science 66:271−282. Partanen, K. 2012. Available in Finnish at: http://mttelo.mtt.fi/toukat-tuottamaan-valkuaisrehua Stein, HH., Sève, B., Fuller, MF., Moughan, PJ., de Lange, CFM. 2007. Invited review: Amino acid bio- availability and digestibility in pig feed ingredients: Termonology and application. Journal of Animal Science 85:172–180. 29 6 Appendix I References for analytical methods used Dry matter DM content was determined by drying at 105C for 16 h. Ash Ash was determined by 600C for 2 h or alternatively 510C 16 h. (AOAC, 1990. Official Methods of Analysis. Association of Official Analytical Chemists, Inc., Arlington, VA. 1298 p. ISBN 0-935584-42-0). Ether extract after hydrolysis with 3M HCl Accredited In-house methods No. 4.21 and 4.22: Determination by Soxcap-Soxtec- Analyzer. (AOAC Official Method 920.39 Fat (Crude) or Ether Extract in Animal Feed and Foss Tecator Application Note AN 390). Nitrogen (Crude protein) by Kjeldahl method Accredited methods 1120, 1122 and 1125 Kjeldahl; Standard methods (AOAC, 1990. Official Methods of Analysis. Association of Official Analytical Chemists, Inc., Arlington, VA. 1298 p. ISBN 0-935584-42-0) using Cu as a digestion cata- lyst and using Foss Kjeltec 2400 Analyzer Unit (Foss Tecator AB, Höganäs, Swe- den). Neutral detergent fibre (NDF) with filtering apparatus by Van Soest, P.J., Robertson, J.B. and Lewis, B.A. 1991. Methods for dietary fi- bre, neutral detergent fibre and nonstarch polysaccharides in relation to animal nu- trition. Journal of Dairy Science, 74: 3583-3597. Sodium sulfite was used in NDF-detergent solution and α-amylase in case of sam- ples containing starch. NDF is expressed without containing residual ash. Acid Detergent fibre (ADF) and Lignin (Permanganate-lignin) by Robertson, J.B. and Van Soest, P.J. 1981. The detergent system of analysis and its application to human foods. In: James, W.D.T. and Theander, O. (eds.). The Analyses of dietary Fibre in Foods. New York, NY, Marcell Dekker. p. 123-158. Amino acids In-house method No. 5000: Determination of amino acids (UPLC). European Commission (1998). Commission Directive 98/64/EC. Community Methods of Analysis for the determination of amino acids, crude oils and fats, and olaquindox in feeding stuffs and amending Directive 71/393/EEC. Official Journal L 257, 19/09/1998 p. 14-28. Total (peptide bound and free) amino acid analysis was performed with Biochrom 20 amino acid analyser (Biochrom Ltd, Cambridge, England) using Sodium Buffer –system. Since 1.1.2009 the equipment used was Waters Finland MassTrak UPLC (Waters Corporation, Milford, U.S.A) and the application was UPLC Amino Acid Analysis Solution®. 30 In virto (pigs), apparent ileal digestibility of N and dry matter by Boisen, S. and Fernàndez, J.A. 1995. Prediction of the apparent ileal digestibil- ity of protein and amino acids in feedstuffs and feed mixtures for pigs by in vitro analyses. Animal Feed Science and Technology, 51: 29-34. In vitro (pigs), total tract digestibility of organic matter by Boisen, S. and Fernàndez, J.A. 1997. Prediction of the total tract digestibility of energy in feedstuffs and in pig diets by in vitro analyses. Animal Feed Science and Technology, 68: 277-286 Minerals and trace elements (Ca, P, K, Na, Mg, Mn, Fe, Cu, Zn, S ) by Luh Huang, C.-Y. and E.E. Schulte. 1985. Digestion of plant tissue for analysis by ICP emission spectrometry. Communications in soil science and plant analysis 16: 943-958. Measurement was performed with ICP-OES (inductively coupled plasma optical emission spectrometry) (Thermo Jarrel Ash Iris Advantage, Franklin, USA). Titanium Digestion of samples for Titan analysis was made according to van Bussel, W., Kerkhof, F., van Kessel, T., Lamers, H., Nous, D., Verdonk, H., and Verhoeven, B. 2010. Accurate determination of Titanium as Titanium Dioxide for limited sample size digestibility studies of feed and food matrices by inductively coupled plasma optical emission spectrometry with real-time simultaneus internal standardization. Atomic Spectroscopy 31 (3): 81-88. Sugars Somogyi, M. 1945. A new reagent for the determination of sugars. Journal of Biological Chemistry 160: 61-68. 31 II The scale for the evaluation of the gastric lesions in pigs (Hautala and Rautiainen 1991) 32 III Photograph of dry Hermetia meal (Photo: Tapio Helenius)